The long-term goals of this proposal are to investigate the antecedents of successful aging, age-related cognitive decline, and dementia in a unique population of 2501 participants in the Baltimore Longitudinal Study of Aging (BLSA), a 40-year prospective study of aging conducted by the National Institute on Aging. In 1994, studies of risk factors for Alzheimer's Disease in this cohort were extended to include antecedents for additional outcomes: a) rate of memory loss with aging; b) successful aging [longevity >90 years, Active Life Expectancy (ALE)]; and c) dementia, particularly AD. By examining psychological differences, detected in AD patients 10-20 years before dementia, Specific Aim 1 is designed to test the hypothesis that subjects with higher cognitive performance (visual memory and vocabulary) in midlife, will have greater longevity and ALE. Specific Aim 2 will investigate family history and genetic markers, initially for Apo E alleles in promoting or preventing cognitive decline. Antecedent physiological measures (including glucose tolerance, insulin secretion, plumonary and cardiac function, and lipid profiles) will be determined for each outcome in Aim 3, and antecedent medical hisory variables are examined in Specific Aim 4. A noncurrent prospective design is imposed on the BLSA subjects. Subjects are traced to the present to determine outcomes in all active, inactive and deceased participants by direct examination (including home visits), medical records, and collateral informant interviews. Cross-sectional and longitudinal analytic strategies are used to examine prospective data up to 40 years before outcome. The BLSA provides a unique database to study the precursors of cognitive aging and longevity. Based on the concept that the processes of aging are lifelong, our studies will provide extraordinarily valuable inofrmation for understanding the biologcial expression of aging in health and disease.